The script handling part of the engine appeared to be too fragile and dependent on
a precise timing of **deleting** a ccInstance.
For example, restoring a saved game may execute in the midst of the call to
RunScriptFunction(), at the "post script cleanup" stage, while there are still references
to ccInstance on the engine's call stack (and few other places). If anything prevents
ccInstance's disposal at this moment, that would in turn lead to logical errors due to
the game scripts "instances" counting. The most outstanding consequence is that the
script exports are kept pointing into the deleted memory and not re-registered after the save is restored.
Overall, the script handling part of the engine would benefit from a major refactor,
with a goal to make it safe and consistent, but this is not something I might do at the moment.
From upstream 1f6abf4b678f77825c710d4ae18b57ff8f882c93
After the updates to cc_instance this should now work
From upstream b1e4f75141ce88b3146f55ecd9d7a95378428072
and 49e8aa0b566dea4ce025f9530a9f2392dbe67a33
In order to specify that they should be treated as signed
Some ports may treat an (unspecified) "char" as unsigned by default. Android does this for its ARM architectures and while we did fix this by enforcing -fsigned-char to its compiler, the issue may come up for other ports. Also it should be better to clarify in the engine when a variable is not actually storing string characters.
I've spotted and changed the most "safe" cases I could find for this. As far as I can tell, out of these, only the Outline field and fontoutline array could potentially be assigned valid negative values which would cause bugs for ports treating chars as unsigned.
Walter Agazzi
Thierry Crozat
Paul Gilbert
Eugene Sandulenko
antoniou79